CoreXY - rotate your parts for better print?



  • I run a quick experiment of printing two calicats on a CoreXY machine, one in normal orientation and the other rotated 45deg around Z.

    Looking carefully with reflected light, the surface of the normal one has tiny vertical lines which the rotated one doesn't.

    My guess is that the rotation causes each rectangular line to be printed with a single motor, avoiding the 'zig-zag' affect of combining two step-wise motions.

    Anybody can reproduce these results?

    My setting: 1.8deg steppers, 20T pulleys, 16 microsteps, 100mm/s, PETG 240C.

    M350 X16 Y16 Z16 E16 I1                      ; Configure microstepping with interpolation
    M92 X80.00 Y80.00 Z400.00 E411.8             ; Set microsteps per mm
    


  • @zapta I don't have your problem on my machine, so changing the orientation of the part on the build plate makes no difference to me. I suspect your issue is caused by some sort of resonance which may change when using two motors rather than one, and that is setting up a sympathetic resonance in your hot end and /or it's mounting arrangement. My hot end is very rigidly mounted and has a relatively huge amount amount of mass, so is largely unaffected by resonances and vibrations.



  • @deckingman, printing the calicat at 100mm/s didn't get the machine into the resonance mode and it sounded more like a quiet whoosh-whoosh which I liked and appreciated. 😉

    Have you run the comparison on your machine and looked closely at the results? e.g. with a rotate and unrotated test cubes.



  • @zapta Yes, many times in the past. As I said in my previous post, I don't have a problem with surface finish, regardless of the orientation on the build plate.

    Edit. I'm not disputing the fact that you have a problem but the fact that my CoreXY does not have the same problem, would indicate that it is something to do with your particular machine, rather than a fundamental issue with CoreXY kinematics.



  • Thanks @deckingman. It was not clear to me from your first post if you actually tested it or just expecting the rotated model to be the same.

    This is my first DIY/CoreXY machine and I am very happy with the print results, speed, and overall simplicity, but will keep tweaking it for a while to see how much I can push the envelope. For example, will try 0.9deg/16T combination to see if it eliminates the occasional high speed vibrations/noise.



  • @zapta said in CoreXY - rotate your parts for better print?:

    Thanks @deckingman. It was not clear to me from your first post if you actually tested it or just expecting the rotated model to be the same.

    This is my first DIY/CoreXY machine and I am very happy with the print results, speed, and overall simplicity, but will keep tweaking it for a while to see how much I can push the envelope. For example, will try 0.9deg/16T combination to see if it eliminates the occasional high speed vibrations/noise.

    For info, I started building my machine about 4 years ago and I'm still tweaking it ☺

    Or more precisely, it has gone through several designs iterations and it continues to evolve - I think it always will.



  • @deckingman said in CoreXY - rotate your parts for better print?:

    For info, I started building my machine about 4 years ago and I'm still tweaking it
    Or more precisely, it has gone through several designs iterations and it continues to evolve - I think it always will.

    isnt that what most of us do? 🙂



  • @veti said in CoreXY - rotate your parts for better print?:

    isnt that what most of us do?

    This is a trap I try to avoid in general, more focus on tools and equipment than on the hobby itself (e.g. bikes vs biking), but I guess I can redefine this hobby as 3D printers rather than 3D printing 😉



  • @zapta said in CoreXY - rotate your parts for better print?:

    0.9deg/16T combination

    I'm curious to see if this has an effect. I suspect it will.

    I wonder also if you'd notice any similar surface artifacts when printing a cylinder, in the areas close to where the motor movement would phase into the same pattern as the 45 degree portions of the cube.

    Also pay attention to your belts and pulleys and how they mesh. I noticed a difference after switching from cheap belts and pulleys to proper gates belts and pulleys. They seemed to mesh together more smoothly causing less deviation. The same can be true of belt teeth riding on smooth idler surfaces.

    After I eliminated these surface artifacts, the next thing that became apparent was noise from the V wheel bearings appearing as light vertical waves that meander up flat wall surfaces. Only visible in certain light and hard to photograph. Visible only, no physical texture to it. Completely impossible to see on more naturally curved surfaces. More visible on low layer heights.

    I guess what I'm saying, is that it's a rabbit hole. Curing the most obvious defects just lets the more subtle ones shine through.

    Video showing artifacts



  • Also pay attention to your belts and pulleys and how they mesh. I noticed a difference after switching from cheap belts and pulleys to proper gates belts and pulleys.

    I do have gates belt on order. The 20T pulleys I use seem to be of low quality and wobbles (subjectively, didn't measure it).

    Where can I get good quality 20T and 16T GT2 pulleys? (for the steppers with set screws and also with and without tooth for the idlers)

    I guess what I'm saying, is that it's a rabbit hole.

    Yes, it's is, but I am almost there, just one more problem to solve 😉



  • E3D is carrying gates belts and pulleys now, so that's a good option.
    Ebay sellers are another option though it's always a gamble.



  • @phaedrux, thanks. Found it at a local E3D distributor.


 

Looks like your connection to Duet3D was lost, please wait while we try to reconnect.